xlnetFv4_ftis_noPretrain

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7658
• Accuracy: 0.4289
• Macro F1: 0.1717

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 53850
• training_steps: 1077000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
3.2778	0.0009	1000	3.4101	0.1492	0.0508
2.6169	1.0009	2000	3.0657	0.3318	0.0747
2.0767	2.0008	3000	2.5978	0.3767	0.0775
1.9754	3.0007	4000	2.4066	0.4042	0.0882
2.0542	4.0006	5000	2.2766	0.4015	0.0917
1.8771	5.0006	6000	2.2322	0.4058	0.1041
1.9728	6.0005	7000	2.1515	0.3971	0.1038
1.9356	7.0004	8000	2.1180	0.4048	0.1076
1.8861	8.0004	9000	2.0644	0.4044	0.1101
1.8417	9.0003	10000	2.0713	0.3948	0.1095
1.9464	10.0002	11000	2.0515	0.3801	0.1081
1.9529	11.0001	12000	2.0508	0.3754	0.1055
1.8001	12.0001	13000	2.0173	0.3900	0.1110
1.895	12.0010	14000	2.0202	0.3920	0.1123
1.9938	13.0009	15000	1.9929	0.3601	0.1032
1.9619	14.0009	16000	1.9595	0.3882	0.1153
1.9526	15.0008	17000	1.9886	0.3566	0.1079
1.792	16.0007	18000	1.8701	0.3935	0.1162
1.7947	17.0006	19000	1.9231	0.4008	0.1267
1.964	18.0006	20000	1.9059	0.4176	0.1343
1.8604	19.0005	21000	1.8850	0.3899	0.1239
1.9679	20.0004	22000	1.8949	0.4385	0.1398
1.9046	21.0004	23000	1.8807	0.3680	0.1262
1.786	22.0003	24000	1.8673	0.4111	0.1437
1.902	23.0002	25000	1.8611	0.4282	0.1429
2.0151	24.0001	26000	1.8620	0.3950	0.1350
1.8067	25.0001	27000	1.8583	0.4070	0.1387
1.7885	25.0010	28000	1.8710	0.4190	0.1427
1.9458	26.0009	29000	1.8955	0.4172	0.1382
1.8265	27.0009	30000	1.8044	0.4128	0.1385
1.8682	28.0008	31000	1.8223	0.3833	0.1313
1.8845	29.0007	32000	1.8387	0.3893	0.1348
1.8141	30.0006	33000	1.8078	0.4263	0.1465
1.9524	31.0006	34000	1.8476	0.4008	0.1424
1.8705	32.0005	35000	1.7960	0.3955	0.1365
1.9233	33.0004	36000	1.8082	0.4222	0.1515
1.7887	34.0004	37000	1.8289	0.4197	0.1392
1.8995	35.0003	38000	1.8094	0.3644	0.1302
1.9193	36.0002	39000	1.7976	0.4105	0.1333
1.807	37.0001	40000	1.7958	0.3994	0.1258
1.6897	38.0001	41000	1.8255	0.4009	0.1400
1.8465	38.0010	42000	1.8031	0.3839	0.1352
1.9148	39.0009	43000	1.8039	0.3886	0.1399
1.9204	40.0009	44000	1.8456	0.3819	0.1312
1.8785	41.0008	45000	1.8408	0.4060	0.1466
1.9642	42.0007	46000	1.8162	0.3727	0.1427
1.9356	43.0006	47000	1.8446	0.4005	0.1513
1.7817	44.0006	48000	1.7895	0.4063	0.1453
1.8289	45.0005	49000	1.7917	0.4138	0.1392
1.8518	46.0004	50000	1.8293	0.4007	0.1589
1.7782	47.0004	51000	1.7812	0.4088	0.1355
1.9633	48.0003	52000	1.8348	0.4056	0.1514
1.9718	49.0002	53000	1.7777	0.4426	0.1595
1.9248	50.0001	54000	1.7779	0.4174	0.1501
1.7681	51.0001	55000	1.7986	0.4014	0.1368
1.8174	51.0010	56000	1.8043	0.4056	0.1420
1.841	52.0009	57000	1.7879	0.3974	0.1425
1.8102	53.0009	58000	1.8208	0.4252	0.1552
1.8059	54.0008	59000	1.7790	0.4062	0.1501
1.8195	55.0007	60000	1.7846	0.4059	0.1538
1.7883	56.0006	61000	1.7771	0.3950	0.1535
1.8632	57.0006	62000	1.7695	0.4177	0.1601
1.9495	58.0005	63000	1.7984	0.4275	0.1601
1.9593	59.0004	64000	1.7376	0.4261	0.1488
1.8409	60.0004	65000	1.7984	0.3857	0.1509
1.8503	61.0003	66000	1.7980	0.3936	0.1593
1.9144	62.0002	67000	1.7780	0.4075	0.1613
1.8632	63.0001	68000	1.8192	0.4245	0.1565
1.7526	64.0001	69000	1.7383	0.4045	0.1525
1.6273	64.0010	70000	1.8099	0.4485	0.1603
1.8939	65.0009	71000	1.7749	0.4344	0.1708
1.8592	66.0009	72000	1.7911	0.3803	0.1374
1.8739	67.0008	73000	1.7427	0.4399	0.1599
1.7345	68.0007	74000	1.8108	0.4179	0.1396
1.9316	69.0006	75000	1.7658	0.4289	0.1717
1.7924	70.0006	76000	1.7827	0.4247	0.1662
1.8339	71.0005	77000	1.7344	0.4266	0.1623
1.9731	72.0004	78000	1.8000	0.3535	0.1539
1.8868	73.0004	79000	1.7762	0.3975	0.1566
1.8885	74.0003	80000	1.7581	0.4000	0.1642
1.8781	75.0002	81000	1.8021	0.3695	0.1360
1.9189	76.0001	82000	1.7375	0.4177	0.1551
1.8382	77.0001	83000	1.8088	0.3697	0.1592
1.828	77.0010	84000	1.7752	0.4315	0.1585
1.8672	78.0009	85000	1.7715	0.4054	0.1684
1.8834	79.0009	86000	1.7985	0.3988	0.1603
1.783	80.0008	87000	1.7518	0.4374	0.1683
1.8679	81.0007	88000	1.7966	0.3770	0.1549
1.8818	82.0006	89000	1.7799	0.4094	0.1673
1.7993	83.0006	90000	1.7827	0.3770	0.1504
1.9272	84.0005	91000	1.7251	0.4290	0.1576
1.8129	85.0004	92000	1.7738	0.3877	0.1580
1.8326	86.0004	93000	1.7855	0.4101	0.1641
1.9804	87.0003	94000	1.7172	0.4276	0.1676
1.814	88.0002	95000	1.8198	0.3801	0.1560

Framework versions

• Transformers 4.46.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.1